Some findings from across our Varick customers that might shape how you think about AI adoption going forward:
1. Customers are getting wiser about spend. A few months ago, most were willing to spend an unlimited amount on tokens from OpenAI and Anthropic. Today they're asking us to diligence their AI spend and to match the right model to the right work. That means more work for us and meaningful savings for them. They want to know AI is actually cheaper than just adding headcount, and they want the math to back it up.
2. Customers are accepting that this isn't instant. A few months ago, customers expected to become AI-native over the course of a week and a few software adoptions. Now they're accepting the reality: becoming AI-native means rethinking the architecture of your entire company. I don't mean that in a corny sense. It literally means changing the org charts, the work, and the handoffs in every crevice of the company.
3. Customers are done with the big shops. Microsoft, IBM, McKinsey, Deloitte, and the rest are all pitching AI transformations, and our customers are fed up with paying eight figures for a slide deck. This is the whole reason we have a business: we sit at the bleeding edge of AI while having the business sense to identify the root of a problem and then build the agents to solve it. This used to be something I had to convince customers of before the first sales call. Now they're the ones telling me, "we're never working with McKinsey again."
This past month we had the highest inbound volume we've ever seen. Between April 1 and May 1, 56 companies doing between $500M and $25B in revenue reached out to us. Some of these companies are direct competitors with one another. It's fascinating to watch the race for enterprise AI-nativity unfold in real time. I like to imagine that we change the course of history for our clients.
AI transformations are now the consensus path to realizing AI ROI. AI SaaS doesn't get you there, and giving every employee a Claude or Cowork subscription doesn't move the needle either. The only way to get to hundreds of millions in annual ROI is to combine a team that can learn your business processes on the ground with the engineering capacity to automate every manual process worth automating.
We bet on this thesis a year ago, and we're vindicated more every single day. If you're interested in transforming your company with AI, or interested in joining the company that's leading the AI transformation wave, visit our website.
Cheers.
George Nimeh retweeted
🚀 Want FREE models you can plug into OpenClaw or Hermes?
Here are 9 resources you can use for free access to model APIs
No local setup, no credit card, just pure cloud APIs with OpenAI-compatible endpoints
You can’t get free Opus quality (yet) but all of these have genuine free tiers right now (rate limits may apply) and are good enough to get started if you don’t want to spend $ to get started with agents
1️⃣ OpenRouter Free Models
(Gemma 4 31B/26B, NVIDIA Nemotron 3 Super 120B MoE, MiniMax M2.5, Qwen3 variants, Llama 4/3.3, gpt-oss-120B, Arcee Trinity, etc.)
• ~29 completely free $0/M token models
• Insane variety + top-tier open model evals (especially coding & agents)
• Best for rotating models automatically
👉 Sign up: openrouter.ai/keys
2️⃣ Google Gemini API
(Gemini 2.5 Pro / Flash series)
• Strongest overall free frontier model
• Excellent multimodal, 1M+ context, native tool calling & agentic performance
• Very generous free limits (often 5–15 RPM)
👉 Sign up: aistudio.google.com/app/apik…
3️⃣ NVIDIA
(Nemotron variants, Llama 3.3 70B, Qwen3 235B, Mistral Large, etc.)
• Optimized high-performance open models
• Free prototyping tier (~40 RPM)
👉 Sign up: build.nvidia.com/explore/dis…
4️⃣ Grok Cloud
(Llama 4 Scout, Llama 3.3 70B, Qwen3 32B, gpt-oss models, etc.)
• Blazing-fast inference (hundreds of tokens/sec)
• Perfect for real-time agents
• Strong open-model performance with solid free tier
👉 Sign up: console.groq.com/keys
5️⃣ Cerebras Cloud
(Qwen3 235B, Llama 3.3 70B, DeepSeek variants, etc.)
• Massive models with excellent reasoning/coding evals
• Very generous daily free limits (~30 RPM, up to 1M+ tokens/day on some)
👉 Sign up: cloud.cerebras.ai/
6️⃣ Mistral La Plateforme
(Mistral Large 3, Small 3.1, Ministral 8B, etc.)
• Strong in coding, multilingual & agentic tasks
• Solid free tier (~1 req/s, ~1B tokens/month)
👉 Sign up: console.mistral.ai/api-keys
7️⃣ Cohere
(Command A, Command R+, Aya Expanse 32B, etc.)
• Free tier: 20 RPM, 1K requests/month
👉 Sign up: dashboard.cohere.com/api-key…
8️⃣ GitHub Models
(Llama 3.3 70B, DeepSeek R1, some GPT-4o previews, etc.)
• Decent mid-tier evals with easy GitHub integration
• Free tier limits (10–15 RPM)
👉 Sign up: github.com/marketplace/model…
9️⃣ Cloudflare Workers AI
(Llama 3.3 70B, Qwen QwQ 32B, etc.)
• Lightweight but solid for simple agents
• Free tier: 10K neurons/day
👉 Sign up: dash.cloudflare.com/profile/…
Pro tips for agent builders:
• Most work instantly with OpenAI SDK (just change base URL + your key)
• Start with OpenRouter for quality/variety (they often feature new free models)
• Add Groq as speed fallback
• Rotate providers when you hit caps
Free intelligence for your agent is just a signup away!
George Nimeh retweeted
Why is no one talking about this?
@nvidia is offering around 80 AI models via hosted APIs absolutely for free.
You get access to MiniMax M2.7, GLM 5.1, Kimi 2.5, DeepSeek 3.2, GPT-OSS-120B, Sarvam-M etc.
This plugs straight into OpenClaude, OpenCode, Zed IDE, Hermes agent and even with Cursor IDE.
Setup:
– Grab API key: build.nvidia.com/models
– base_url = "integrate.api.nvidia.com/v1"
– api_key = "$NVIDIA_API_KEY"
– select model (e.g. minimaxai/minimax-m2.7)
If you’re building or experimenting, this is basically free inference.
Lock in and start building today anon.
Thank me later.
George Nimeh retweeted
Replying to @sama
sam just so you know for next time this is technically incorrect usage of the “-maxx” suffix
to “looksmaxx” is to improve oneself, it’s not really a comparison between two different things (“-mogged” is a comparison however). so the correct usage here would’ve been something like:
“we still get looksmogged on frontend…”
or
“we still need to looksmaxx our frontend…”
George Nimeh retweeted
Claude is not allowed to write outside the workspace.
But it wanted to.
So Claude wrote a python script and executed it via bash to modify the file essentially hacking my permissions.
George Nimeh retweeted
Anthropic generated enormous attention for their “we’re the good guys" ruse and even got famed AI model technology expert Christine Lagarde to chime in.
Oh look! Anthropic's entire "we are delaying Mythos" narrative was marketing hogwash.
Kudos to FT for confirming what was obvious. Anthropic simply doesn't have the compute.
FT: "Multiple people with knowledge of the matter suggested Anthropic was holding back from a wider release until it could reliably serve the model to customers."
George Nimeh retweeted
JD Vance is lecturing the Pope on Catholicism and Pierre Poilievre is lecturing Mark Carney on economics and RFK Jr is lecturing scientists about vaccines and Donald Trump is lecturing the world on tariffs and Pete Hegseth is quoting Pulp Fiction and thinking it’s the Bible
George Nimeh retweeted
Dario is wrong.
He knows absolutely nothing about the effects of technological revolutions on the labor market.
Don't listen to him, Sam, Yoshua, Geoff, or me on this topic.
Listen to economists who have spent their career studying this, like @Ph_Aghion , @erikbryn , @DAcemogluMIT , @amcafee , @davidautor
George Nimeh retweeted
Vancouver, it’s been a blast! 🇨🇦
“The lobster is loose, and it’s not going back into the tank,” says @openclaw founder @steipete. In this brand new talk from #TED2026 he shares why AI agents — built by you — are the future: t.ted.com/DPASxmF
George Nimeh retweeted
In 2007 Gawker outed Peter Thiel as gay.
In 2016 Thiel secretly funded $10M in lawsuits that bankrupted them.
This week he funded the AI that grades journalists.
Here's what's inside it:
1/ It's called Objection. The founder is Aron D'Souza, the same lawyer who personally ran the Gawker takedown for Thiel.
2/ Backers include Peter Thiel and Balaji Srinivasan.
3/ Anyone pays $2,000 to trigger a "public investigation" of any reporter.
4/ The "investigators" are former FBI, NSA, and CIA officers.
5/ The jury is a panel of AIs from OpenAI, Anthropic, Google, Mistral, and xAI.
6/ The output is an "Honor Index" score on the journalist.
7/ Anonymous whistleblowers are ranked dead last in the evidence weighting.
8/ Corporate emails and government filings are ranked at the top.
9/ What used to take 5-10 years in court, they say, now takes 72 hours.
The man who spent a decade using the legal system to destroy a newsroom just automated the process.
Surveillance-state veterans score the reporter. AI trained by Thiel and Musk renders the verdict. Whistleblowers count least. Power's paper trail counts most.
Speech for me. A score for thee.
George Nimeh retweeted
woke up and my mentions are full of these
Both me and @davemorin tried to talk sense into Anthropic, best we managed was delaying this for a week.
Funny how timings match up, first they copy some popular features into their closed harness, then they lock out open source.
George Nimeh retweeted
Claude code source code has been leaked via a map file in their npm registry!
Code: pub-aea8527898604c1bbb12468b…
George Nimeh retweeted
Replying to @shoucccc
What is potentially concerning in this story?
First, the reverse engineering of the permissions system. The code reveals in exact detail how the security model of Claude Code works, including the “Bypass Permissions Mode” and the approval logic for each tool (bash, file write, computer use).
An attacker who understands this architecture can craft more sophisticated prompts or configurations to try to bypass the guardrails, especially in corporate environments where Claude Code runs with elevated permissions.
Second, the complete system prompt is in the code. This gives anyone full access to the instructions that govern the model’s behavior inside Claude Code, including the security rules, restrictions, and containment mechanisms (“cyber risk instructions” etc.).
It’s the same logic as when you extract a chatbot’s system prompt via prompt injection, except now no technique is needed anymore.
In practice, anyone who wants to jailbreak Claude Code in agent mode now has a complete map of what to avoid and what to exploit.
Third, the telemetry. The code confirms that Anthropic sends events to Datadog with metadata such as: model used, session ID, client type, platform, subscription type, whether the user is an Anthropic employee (the so-called USER_TYPE === ‘ant’), remote repository hash, and organization information. It doesn’t send code or file paths (there’s a typing I_VERIFIED_THIS_IS_NOT_CODE_OR_FILEPATHS that acts as an engineering lock to prevent this), but the exact extent of what is collected is now transparent for anyone to audit, and this will inevitably generate privacy debates.
Fourth, the internal infrastructure becomes visible. Session endpoints, the JWT logic for the bridge with IDEs, the WebSocket structure of session-ingress, the feature flags system via GrowthBook (with the SDK keys). None of this compromises user data, but it gives a technical adversary a deep understanding of the platform’s attack surface.
Anthropic will probably rotate the client-side keys as a precaution. Telemetry can be disabled (the code shows: DISABLE_TELEMETRY=1 or CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC=1), which is useful information that is now transparently documented.
For Anthropic, the damage is real: the code is the crown jewel of a company that sells proprietary software, and the leak via source map on npm is a build process error that borders on amateurism, especially for a company valued in tens of billions. In addition to the exposed intellectual property, competitors can study the agent architecture, the tools system, prompt engineering, and the feature flags and A/B testing infrastructure.
In simpler words: heads will roll!
George Nimeh retweeted
🚨⚡️“Americans are very lucky, because wherever they go to bring freedom… they find oil.” - Michele Serra, Italian journalist
George Nimeh retweeted
Most tech podcasts are so fucking boring.
The host is either some person who couldn't hack it in a operating role (VC, engineer, founder, etc.) or some VC who is just doing it to get deal flow.
No one cares what Bill Gurley thinks about AI, they want to hear about why he fucked over TK at Uber.
No one cares what Chamath thinks about politics, they want to hear about what happened at Social Capital and why he closed the fund.
No one cares what Keith Rabois has to say about Miami, they want to hear about what happened at OpenDoor.
No one cares what Sam Altman thinks about AGI, they want to hear about why he was fired from YC and (initially) OpenAI.
But instead we get these boring, sanitized conversations about why SaaS companies are or are not dead.
Topics no one gives us a shit about.
George Nimeh retweeted
Apple understands that users miss the experience of always inserting USB cables wrong the first time, and wondered how they could recreate this experience with USB-C…
We just passed React on GitHub stars. 🦞
Let that sink in. A personal AI assistant built by a lobster-obsessed Austrian and an army of crustacean enthusiasts just outstarred the library that powers half the internet.
We shipped 90+ changes today. They shipped a conference.
George Nimeh retweeted
nothing on earth is more in immediate danger than the stuffed animal they handed nathan mackinnon































